SEARCH

SEARCH BY CITATION

Keywords:

  • Scalp tissue;
  • Multipotent stem/progenitor cells;
  • Neurogenic differentiation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Recent studies have shown that adult tissues contain stem/ progenitor cells capable of not only generating mature cells of their tissue of origin but also transdifferentiating themselves into other tissue cells. Murine skin-derived precursor cells, for example, have been described as unique, nonmesenchymal-like stem cells capable of mesodermal and ectodermal neurogenic differentiation. Human-derived skin precursors are less well characterized.

In this study, the isolation and characterization of adherent, mesenchymal stem cell–like cells from human scalp tissue (hSCPs) are described. hSCPs initially isolated by both medium-selection (ms-hSCPs) and single-cell (c-hSCPs) methods were cultured in medium containing epidermal growth factor and fibroblast growth factor-β. Cultured ms-hSCPs and c-hSCPs demonstrated a consistent growth rate, continuously replicated in cell culture, and displayed a stable phenotype indistinguishable from each other. Both hSCPs expressed surface antigen profile (CDw90, SH2, SH4, CD105, CD166, CD44, CD49d-e, and HLA class I) similar to that of bone marrow mesenchymal stem cells (BM-MSCs). The growth kinetics, surface epitopes, and differentiation potential of c-hSCP cells were characterized and compared with BM-MSCs. In addition to differentiation along the osteogenic, chondrogenic, and adipogenic lineages, hSCPs can effectively differentiate into neuronal precursors evident by neurogenic gene expression of glial fibrillary acid protein, NCAM, neuron filament-M, and microtubule-associated protein 2 transcripts. Therefore, hSCPs may potentially be a better alternative of BM-MSCs for neural repairing, in addition to their other mesenchymal regenerative capacity. Our study suggests that hSCPs may provide an alternative adult stem cell resource that may be useful for regenerative tissue repair and autotransplantations.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Stem cells isolated from developing tissues can be differentiated into more than one specific cell type [1]. The proportion of these multipotent stem cells diminishes with the maturation of the tissue. On the other hand, recent studies have shown that adult tissues contain residual tissue stem/progenitor cells capable of not only generating mature cells of their own tissue but also renewing other tissue cells. A growing body of evidence has also shown that the process of tissue repair is driven by these stem-like cells residing in different tissues. Among candidates for reparative cells are the stem cells from adult bone marrow referred to as either mesenchymal stem cells or marrow stromal cells (BM-MSCs). BM-MSCs are defined by their ability to differentiate into cells of osteogenic, chondrogenic, adipogenic, and, more or less, myogenic lineages. Besides bone marrow [2, 3], the peripheral blood [3], the retina [4], the adipose tissue [57], and the central nervous system [8, 9] are known reservoirs for multipotent, mesenchymal stem cells (MSCs). The adherent, spreading morphology of MSCs is visibly distinct from hematopoietic stem cells. Our current interest is focused on identifying new sources of MSCs and characterizing their potentials for differentiating into nonmesenchymal tissues.

Recently, BM-MSCs were found to undergo neuronal differentiation when they were cocultured with brain tissue in the absence of hematopoietic stem cells [10]. The study implied that a population of cells with neuroectodermal potential can be derived from marrow stroma, the mesodermal mesenchyme origin. Successively, several groups have reported that circulating blood is a reservoir of multipotent MSCs that can be directed into adipogenic, osteogenic, myogenic, neurogenic, hepatogenic or epithelial, endothelial, hepatogenic lineages [3, 1115].

In vitro characterization and maintenance of tissue stem/ progenitor cells is critical to the assessment of their potential for clinical applications. Murine skin-derived progenitors (mSKPs) obtained during developmental stages showed osteogenic, adipogenic, smooth muscle, and neuronal differentiation potentials [16], suggesting that they could be exploited as an alternative source for treating mesenchymal and neurodegenerative disorders. Here we describe that the human scalp tissue contains stem/progenitor cells with mesenchymal and neurogenic differentiation potentials. We isolated the scalp-derived adherent cells (hSCPs) by both medium-selective (ms-hSCPs) and clongenic (c-hSCPs) cultures, characterized their growth kinetics, mesenchymal differentiation potentials, and expression of cell markers, and concluded their characteristics as neurogenic mesenchymal stem/progenitor cells.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Cell Culture

hSCPs were obtained from human scalp tissues from healthy adult patients (35–55 years old) undergoing cosmetic plastic surgery. Samples were collected with patients' consent and with the approved protocol from the human tissue usage committee of Wan-Fang medical center. Briefly, two to three pieces of scalp debris ranging between 4 and 9 mm were postoperatively collected and washed two to three times with phosphate-buffered saline (PBS). The fat tissue was carefully removed, and the scalp was cut into smaller pieces. They were then transferred into a 15-ml centrifuge tube and washed three times to remove any remaining hair or blood clots. Washed tissue pieces were digested with 0.1% trypsin-EDTA (Sigma, St. Louis, http://www.sigmaaldrich.com) in a 37°C water bath for 60 minutes. Digestion was stopped with serum-containing buffer. Next, the scalp pieces were mechanically dissociated in long-term culture medium (Dulbecco's modified Eagle's medium low glucose [DMEM-LG] [GIBCO, Grand Island, NY, http://www.lifetech.com], 10% fetal bovine serum [FBS] [Hyclone, Logan, UT, http://www.hyclone.com], 20 ng/ml epidermal growth factor [EGF], and 20 ng/ml fibroblast growth factor [FGF]-β [R&D Systems, Minneapolis, http://www.rndsystems.com]), and the suspended portion was poured through a cell strainer to remove aggregates and dead cells. Suspended cells were centrifuged at 170g and resuspended in fresh culture medium. These cells were seeded in 10-cm noncoated dishes and incubated in a 5% CO2, 37°C tissue-culture incubator for 4 days. Cells were continuously depleted of contaminating epithelial cells until passage 4, and adherent cells were collected as medium-selected hSCPs (ms-hSCPs) for analyses. ms-hSCPs displayed a uniform cell morphology and were replated at a density of 5 × 104 to 1 × 105 per 10-cm dish after reaching 70%–80% confluence, and 50% of medium was changed every 3–4 days. ms-hSCPs were isolated and cultured with consistent growth rate in a long-term culture medium containing EGF and FGF-β.

The serial dilution method was used to generate single-cell clongenic culture. Briefly, 104 isolated hSCPs were suspended in 1 ml culture medium. In conducting the serial dilution, 100 μl of the cell suspension was transferred into a new eppendroff tube and supplemented with 900 μl of culture medium to make the cell concentration equivalent to 103 cells in 1 ml culture medium. This process was repeated to achieve a final dilution of 10 cells in 1 ml medium. For single-cell culture, 100 μl of the diluted cell suspension was transferred into each well of a 96-well plate containing 200 μl of culture medium. The resultant serial dilution was examined under an inverted microscope (Olympus, Toyko, http://www.olympus.com). Wells with more than one cell or no cell were marked and exempted from the culture selection. Only those wells that contained a single, viable cell were transferred into a larger culture dish to prevent the cell-cell physical contact for developing the clongenic hSCPs (c-hSCPs).

For comparison, commercial human BM-MSCs (BioWhittaker, Walkersville, MD, http://www.cambrexbioproductseurope.com) were cultured in the same basal medium used to culture hSCPs but without cytokines. The cell expansion and long-term culture procedures were adapted for the current hSCP study.

Flow Cytometry Analysis

Scalp-derived adherent cells were isolated, expanded, and characterized by flow cytometric analysis for specific surface antigens. Harvested cells were collected and treated with 0.1% trypsin-EDTA. Cells were stained with fluorescein isothiocyanate– or phycoerythrin-conjugated anti-marker monoclonal antibodies (mAbs) in 100 μl phosphate buffer using titers for 15 minutes at room temperature or 30 minutes at 4°C, as suggested by the manufacturer. Cell surface markers included hematopoietic lineage markers (CD34, CD38, CD45), matrix receptors and angiogenic markers (CD31, CD105), adhesion integrins (CD44, CD49d, CD49e, CD49f, CD106, CD166), factor receptors (EGFR, PDGFR), endothelial progenitor/precursor lineage markers (CD105, CD31), and MSC markers of SH2, SH4, STRO-1, and CDw90. Cells were analyzed using a flow cytometry system (FACSCalibur; Becton, Dickinson and Company, Mountain View, CA, http://www.bd.com). Positive cells were counted and compared with the signal of corresponding immunoglobulin isotypes. Sources of monoclonal antibodies are listed in Table 1.

Table Table 1.. Immunophenotyping of c-hSCPs, ms-hSCPs, and BM-MSCs
  1. a

    Flow cytometry analysis of c-hSCPs, ms-hSCPs, and BM-MSCs. + indicates more than 50% of the population was positive compared with isotype control; ± indicates the positive population ranged within 20%–50%; − indicates less than 20% of the population showed a signal shift compared with isotype control. Sources of monoclonal antibody were as follows: A, Ancell Co., Bayport, MN, http://www.ancell.com; AB, Anawrahta Biotech, Taipei, Taiwan, http://www.anabio.com.tw; B, BD Bioscience; D, Developmental Studies Hybridoma Bank, Iowa City, IA, http://www.uiowa.edu/∼dshbwww; S, Serotec Co., Oxford, U.K., http://www.serotec.com. Isotype anti-IgG1–fluorescein isothiocyanate and anti-IgG1–phycoerythrin monoclonal antibodies were purchased from Serotec Co. Abbreviations: BM-MSC, bone marrow mesenchymal stem cell; c-hSCP, single-cell scalp-derived clongenic stem cell; ms-hSCP, medium-selection scalp-derived stem cell.

 c-hSCPsms-hSCPsBM-MSCsAntibodies
Hematopoietic
CD34B
CD38B
CD45B
HLA class I+++S
Adhesion
CD44+++S
CD49d+++B
CD49e+++B
CD49f+++B
CD106±D
CD166+++B
Angiogenic
CD31S
CD105+++A
Mesenchymal
CD29+++B
SH2+++AB
SH4+++AB
CDw90+++S
Stro-1D
Receptor
EGFR±±±B
PDGFRα+++B

Induction of Osteogenic, Adipogenic, and Chondrogenic Differentiation

In vitro differentiations of MSCs were made in DMEM-LG (GIBCO) supplemented with 10% FBS (Hyclone) and treated with one of the following formulas: adipogenic differentiation medium: DMEM-LG supplemented with 10% FBS, 0.5 mM isobutyl-methylxanthine, 1 μM dexamethasone, 10 μM insulin, and 200 μM indomethacin; osteogenic medium: DMEM-LG supplemented with 10% FBS (Hyclone), 0.1 μM dexamethasone, 50 μM ascorbate-2-phosphate, and 10 mM β-glycerolphosphate; and chondrogenic medium: DMEM-LG supplemented with 1% FBS, 6.25 μg/ml insulin, 10 ng/ml TGF-β1 (R&D), and 50 nM ascorbate-2-phosphate. Drugs used above for differentiation were purchased from Sigma. For osteogenic and adipogenic differentiation, the cell density was 3 × 104 cells per cm2. For chondrogenic differentiation, a higher cell density of 1 to 2 × 105 per 10 μl was used for chondrosphere formation. Medium was changed every 3–4 days for all differentiation assays, and cells were fixed for histochemical staining after 14 days of differentiation.

Calcium Incorporation Assay

Calcium quantitation was carried out following the manufacturer's instruction on the CALCIUM liquicolor kit (lot H047; Human GmbH, Wiesbaden, Germany; http://www.human.de). Briefly, the working reagent was prepared by adding RGT (Color Reagent, 8-hydroxyquinoline, 14 mM; o-cresolphthalein complexone, 0.1 mM; hydrochloric acid, 40 mM) to BUF (buffer solution, lysine buffer, pH 11.1, 0.2 M; sodium azide 0.095%) in 1:1 ratio and allowed to stand for 10 minutes at room temperature before use. Briefly, undifferentiated MSCs at 105 cells were collected in 1 ml 1 × PBS and cultured in the osteogenic medium for 14 days. After removal of the culture medium, the adhesive cells and extracellular matrix were collected by 0.025% trypsin treatment and the final volume was made to 1.0 ml. A sample volume of 20 μl was pipetted into cuvettes and mixed. The absorbance was measured against the blank within 5–50 minutes according to the manufacturer's instruction.

Neurogenic Differentiation of MSCs

MSCs at passages 4 through 6 were incubated in routine culture medium (10% FBS + 1 × AA + 20 ng/ml EGF + 20 ng/ml FGF-β in DMEM-LG medium) for at least 3 days as a preinduction condition. After the preinduction step, cells were treated with the neurogenic differentiation NC210 medium. NC210 consisted of 10% FBS, 2% dimethylsulfoxide (Sigma), 200 mM butylated hydroxyanisole (Sigma), 25 mM KCl (Sigma), 2 mM valproic acid (Sigma), 10 μm forskolin (Sigma), 1 μm hydro-cortisone (Sigma), 5 μg/ml insulin (Sigma), 0.5 mM isobutyl-methylxanthine (Sigma), and 1 mM cAMP (Sigma) in DMEM-LG medium.

After induction, cells were analyzed by immunohistostaining for neuron-specific enolase (NSE) protein (Chemicon, Temecula, CA, http://www.chemicon.com) and nucleic DAPI (Molecular Probes, Eugene, OR, http://www.probes.com) or reverse transcription–polymerase chain reaction (RT-PCR) for neuronal genes.

RT-PCR Analysis of Neurogenic Gene Expression

Expression of microtubule-associated protein 2 (MAP-2) (neuronal lineage), neuron cell-adhesion molecules (NCAMs) (neuronal lineage), neuron filament-M (NF-M) (neuronal lineage), and glial fibrillary acid protein (GFAP) mRNAs was examined by RT-PCR. Total cellular RNA was isolated using RNeasy total RNA isolation kit (Qiagen, Inc., Valencia, CA, http://www.qiagen.com), and cDNA was synthesized using the SuperScript First-strand Synthesis System (Life Technologies, Barcelona, Spain, http://www.invitrogen.com). Specific genes were amplified by PCR using Fast-Run Taq Master Kit (Protech Technology, Taipei, Taiwan, http://www.bio-protech.com.tw). The primer sequences used for amplification of microtubule-associated protein 2 (MAP2), GFAP, and internal control human beta-actin were as follows: MAP2 sense strand, 5′-CTGTCCCTAGGTCAGCTTGC-3′; antisense strand, 5′-GCATGGTGGCTCCCAATCTAT-3′; GFAP sense strand, 5′-TGCCATCTTGGTGCCGA-3′; anti-sense strand, 5′-CTTGACATTACCACCTCCAGGT-3′; NCAM sense strand, 5′-CTCGGCCTTTGTGTTTCCAG-3′; antisense strand, 5′-TGGCAGGAGATGCCAAAGAT-3′; NF-M sense strand, 5′-CTTCAGCCAGTCCTCGTCCC-3′; antisense strand, 5′-TCCTCCAGGTGGTCCGAGTC-3′; hSOX2 sense strand, 5′-CAAGATGCACAACTCGGAGA-3′; antisense strand, 5′-GTTCATGTGCGCGTAACTGT-3′; beta-actin sense strand, 5′-GTGGGGCGCCCCAGGCACCA-3′; antisense strand, 5′-CTCCTTAATGTCACGCACGATTTC-3′. The cDNA product was amplified by PCR using standard methods.

Histochemical Staining

After 14 days of cell culture, media for all differentiation assays were removed from the culture dishes and cells were washed twice with PBS. Cells were fixed in 4% paraformaldehyde (Sigma). For osteogenic differentiation, the cells cultured in osteogenic differentiation medium were stained with von Kossa (Sigma) to identify extracellular matrix calcium mineralization. Briefly, the cells were rinsed with distilled water and then incubated in a 1% (wt/vol) silver nitrate (Sigma) solution in the dark for 30 minutes. They were washed several times with distilled water and developed under light for 60 minutes. For adipogenic differentiation, cells cultured in adipogenic differentiation medium were stained with Oil Red O reagent (Sigma) to examine oil droplet generation in cytoplasm. The cells were incubated in 2% (wt/vol) Oil Red O reagent for 5 minutes at room temperature. Excess stain was removed by 70% ethanol, followed by several washes in distilled water. The cells were counterstained for 2 minutes with hematoxylin (Sigma). Cells cultured in chondrogenic differentiation medium were stained with Safranin-O and Alcian Blue (Sigma) to confirm and locate the proteoglycans and acidic mucopolysaccharides, respectively. 1% Alcian Blue in 0.1 N HCl was used to stain cells at room temperature for 5 minutes and washed with 0.1 N HCl to remove excess dye.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Cell Isolation and Characterization

The total amount of cells isolated from each scalp tissue sample ranged from 5 × 104 to 1 × 105 cells. Approximately 0.5%–2% of the isolated tissue cells was found to be the adherent hSCPs. Morphologically, the ms-hSCP or c-hSCP cells were less flattened and fibroblast-like compared with BM-MSCs (Figs. 1A, 1B). Growth rate for c-hSCPs and BM-MSCs was comparable, as shown in Fig. 1C. The doubling time of ms-hSCPs and c-hSCPs derived from human scalp was approximately 40 hours, whereas that of BM-MSCs was approximately 48 hours. Similarities of hSCP morphology and growth kinetics were observed in more than 30 tissue isolations (one patient per isolation). No significant difference of growth rate, cell frequency, and doubling time was observed among these scalp tissue isolations (data not shown).

thumbnail image

Figure Figure 1.. Morphology and cell growth rate of scalp-derived clongenic stem cells (c-hSCPs) (A) and bone marrow–derived mesenchymal stem cells (BM-MSCs) (B). Representative micrographs (A, B) and growth rates of c-hSCPs (▴) and BM-MSCs (▪) are illustrated (C). After day 35, both c-hSCPs and BM-MSCs continuously proliferated until the cell size and growth rate became large and slow.

Download figure to PowerPoint

Both c-hSCPs and ms-hSCPs expressed an immune-phenotype antigen profile similar to BM-MSCs and shared common cell markers, as shown in Table 1 and Figure 2, including the following: MSC: CDw90, SH2, and SH4 but not Stro-1; hematopoietic: CD45−, CD38−, and CD34−; HLA: class I+; hyaluronate receptor: CD44+; integrins: CD49d+, CD49e+, CD49f+; ALCAM: CD166+; angiogenic: CD105+, CD31−; and cytokine receptors: EGFR+, PDGFRα+. Notably, a slightly higher frequency of cells of hSCPs expressed PDGFRα and CD49f antigens compared with the BM-MSCs.

thumbnail image

Figure Figure 2.. Fluorescence-activated cell sorting analysis of scalp-derived clongenic stem cell surface markers.

Download figure to PowerPoint

Mesenchymal Differentiation of c-hSCPs and ms-hSCPs

We observed that c-hSCPs and ms-hSCPs exhibit similar mesenchymal differentiation capacity as BM-MSCs do.

Osteogenesis

To examine whether hSCPs were as capable of mesenchymal differentiation as the BM-MSCs, c-hSCPs and ms-hSCPs were subjected to osteogenic lineage differentiation. After 1–2 weeks of differentiation in culture, osteogenic-like cell structure gradually formed and stained positive for von Kossa staining. The calcium phosphate formed in these cells precipitated along the cell membrane and showed up as brown, large, aggregate particles embedded in the extracellular matrix when stained with von Kossa (Fig. 3).

thumbnail image

Figure Figure 3.. Osteogenic differentiation of single-cell scalp-derived clongenic stem cells (c-hSCPs), medium-selection scalp-derived stem cells (ms-hSCPs), and bone marrow mesenchymal stem cells (BM-MSCs). After 14 days, a calcified extracellular matrix (arrow, dark black area) was present and positive for von Kossa staining.

Download figure to PowerPoint

In the calcium incorporation assay, the amount of calcium phosphate precipitation increased from 2.38 μg per 105 cells to 20 μg per 105 cells in c-hSCP osteogenesis, whereas in BM-MSCs osteogenesis, it increased from 0.01 μg per 105 cells to 7 μg per 105 cells. Alkaline phosphate staining produced similar results (data not shown). We therefore concluded that c-hSCPs were capable of similar osteogenic differentiation as the BM-MSCs.

Chondrogenesis

Chondrogenic differentiation was achieved by dropping high-density (5 × 105 to 1 × 106) c-hSCPs and ms-hSCPs or BM-MSCs on the dish center containing chondrogenic differentiation medium. The cell condensed and formed chondrosphere-like pellets in 3 days. Frozen dissected specimens were collected after 2 weeks in differentiation culture. Alcian Blue staining was used to visualize mucopolyglycan formation in histological dissections of the above cultures. Mucopolyglycan was present in cartilaginous matrices and lacunae with extracellular proteoglycan formation (Fig. 4). In addition, positive Safranin-O staining was observed for condensed sulfate proteoglycans formation in both BM-MSC and hSCP cultures (Fig. 4).

thumbnail image

Figure Figure 4.. Chondrosphere formation in single-cell scalp-derived clongenic stem cells (c-hSCPs), medium-selection scalp-derived stem cells (ms-hSCPs), and bone marrow mesenchymal stem cells (BM-MSCs). Typical chondrosphere formation was observed for c-hSCPs, ms-hSCPs, and BM-MSCs. The acidic mucopolysaccharide was stained with Alcian blue, whereas proteoglycans were stained with Safranin-O. c-hSCPs, ms-hSCPs, and BM-MSCs were positive for both assays.

Download figure to PowerPoint

Adipogenesis

Small oil droplets were observed to appear gradually in the cytoplasm after 1 week of adipogenic induction in both BM-MSC and hSCP cultures. Adipocyte characteristics were confirmed by positive staining with Oil Red O after 2 weeks of induction but to a lesser extent in hSCPs compared with BM-MSCs (Fig. 5).

thumbnail image

Figure Figure 5.. Adipogenic differentiation of single-cell scalp-derived clongenic stem cells (c-hSCPs), medium-selection scalp-derived stem cells (ms-hSCPs), and bone marrow mesenchymal stem cells (BM-MSCs). Cells with the same passage numbers, as in previous assays, were used. Intracellular oil droplets formed and were stained red by Oil Red O for c-hSCPs, ms-hSCPs, and BM-MSCs (arrow, red area).

Download figure to PowerPoint

Neuronal Differentiation Potential of hSCPs

Both hSCPs and BM-MSCs exhibit more or less capability for differentiation along all three mesenchymal lineages. We wondered whether hSCPs display a neurogenic differentiation potential like the reported rodent SKPs [16]. As predicted, hSCPs exhibited significant neurogenic differentiation potential, because more than 80% of the viable c-hSCP displayed morphological changes into contracted cell bodies and elaborate processes (Figs. 6A, 6B) when cultured in a neurogenic induction media. Immunofluorescence staining of NSE provided additional evidence for neuron-differentiating characteristics (Fig. 6C). RT-PCR analysis of neurogenic and early gene expression in c-hSCP culture showed upregulation of MAP2 and NF-M and downregulation of GFAP and early progenitor hSOX2 gene (Fig. 7). These results demonstrated that c-hSCPs are indeed capable of differentiation into neurogenic precursors more efficiently than the BM-MSCs are.

thumbnail image

Figure Figure 6.. Neurogenic differentiation of single-cell scalp-derived clongenic stem cells (c-hSCPs) on poly-L-lysine–coated dishes. Cell morphology before differentiation (A) and after 48 hours of induction in NC210 neurogenic culture media (B). Neurogenic neuron-specific enolase (NSE) expression was confirmed by fluoro-NSE monoclonal antibody (mAb) staining (C). Top panels show c-hSCPs (P4) differentiated into NSE+ neurosphere-like cells (>80%) at various parts of the culture dish after 5 days of induction in NC210 culture condition. Bottom panels show the same cells counterstained by nucleic DAPI as contrasts. Neurogenic differentiation of bone marrow mesenchymal stem cells (30%) confirmed by fluoro-NSE mAb staining as a comparable study (D).

Download figure to PowerPoint

thumbnail image

Figure Figure 7.. Neurogenic-specific genes microtubule-associated protein 2 (MAP2), NCAM, neuron filament-M (NF-M), glial fibrillary acid protein (GFAP), and the early progenitor hSOX2 gene were examined by reverse transcription–polymerase chain reaction analyses. Upregulation of neuronal-specific MAP2 and NF-M and downregulation of glial GFAP and early progenitor hSOX2 gene expressions were detected after neuronal-induction treatment of the single-cell scalp-derived clongenic stem cells (c-hSCPs) in NC210 medium. Target genes were visible at 18 cycles (beta actin), 25 cycles (NCAM, MAP2, and NF-M), and 30 cycles (GFAP and hSOX2), respectively.

Download figure to PowerPoint

Interestingly, GFAP transcription can be detected in the untreated hSCPs but disappeared after subsequent neurongenic differentiation under NC210 culture treatment. Untreated BM-MSCs failed to show detectable GFAP transcripts, although to a lesser extent they can also be differentiated into neuron precursors. We suspected that c-hSCPs behave as a multipotent neuron progenitor cell that could differentiate into both neuron and glial and mesenchymal lineages depending on their chemical treatment, because GFAP transcripts of hSCPs can be detected under different culture conditions (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

The hSCPs

Jahoda's groups have isolated and generated clonal papilla and sheath cell lines from skin tissues and showed that these cells can be transdifferentiated into adipocytes and osteocytes [17]. The scalp skin is known to be composed of precusor cells primarily in hair follicles and basal dermal layers. It is therefore reasonable to entertain the possibility that more than one type of stem cell population can be isolated and expanded by distinct methods.

In this study, the adherent, multipotent MSCs were our initial target in scalp tissue. Our primary cell culture conditions were made selectively for neurogenic differentiation of MSCs, so that epithelial progenitors or mature cells were not likely to survive after four passages. In our experiment, adherent cells expressing markers similar to those of MSCs were reproducibly isolated and expanded, suggesting that adult scalp skin tissue contains multi-potent stem cells that can be obtained by means of either medium-selective or single-cell clongenic culture. Similar cell morphology and growth kinetics were observed for more than 30 separate isolations (one patient per isolation). No correlation between patient age and cell frequency or doubling time was observed (data not shown). These cells are EGF- and FGF-dependent but did not require neuron growth supplement B-27 or leukemia inhibitory factor for short-term cell expansion or to retain multipotency.

hSCPs exhibit cell doubling consistently in the presence of cytokines, long-term expression of MSC surface markers, and extensive, multiple differentiation capacity. hSCPs were positive for fibronectin expression and displayed MSC capacity similar to the well-characterized BM-MSCs [14, 18]. The hSCPs can be induced to differentiate into neuronal lineages expressing glial- and neuron-specific genes, as verified by protein or gene amplification analyses.

The frequency of the cell transdifferentiation potential has been proposed to be located in the hair follicles [17]. Cytokine dependency of hSCPs for survival during short-term culture was distinct from BM-MSCs. It is unknown which part of the dermal tissue harbors this novel and rich population of stem cells in the scalp tissue. Lineage differentiation analysis of hSCPs indicated that scalp tissue contains multipotent stem cells with the potential to differentiate along mesodermal and ectodermal lineages, similar to stem cells found in dermis tissue [17]. Several properties of the hSCP characteristics may imply their important role in wound healing and tissue-repairing potentials.

Differences Between hSCPs and BM-MSCs

In addition to having a pattern of cell morphology, growth rate, adhesion molecules, and mesenchymal cell antigen expression similar to that found in BM-MSCs, hSCPs also shared the most important characteristic of MSCs, namely, the capacity to differentiate into osteogenic, chondrogenic, and adipogenic cell types. In in vitro chondrogenesis, for example, the speed of formation (within 24 hours) and the size of chondrospheres were comparable between hSCPs and BM-MSCs. However, the cytokine dependency of these two cell types was different. We observed a lag-phase doubling time of hSCPs when compared with BM-MSCs grown in cytokine-depleted media, and the cell shape of hSCPs became larger, flattened, and extended, symbolizing their entry into an aging and apoptotic stage (data not shown).

A subpopulation of MSCs has been identified as nestin-positive neurogenic precursors before any neurogenic treatment [16]. We have observed such pre-existing, neurogenic transcripts in routine hSCP culture. On the other hand, in BM-MSC cultures, cells remained at an undifferentiated state before induction of differentiation. Because the mesenchymal differentiation capacities and efficiency between hSCPs and BM-MSCs were comparable, the pre-existing neurogenic transcripts found specifically in hSCPs before any cytokine treatment implied that the varied environmental cues in scalp are different from those found in bone marrow.

hSCPs Exhibited Neuron Stem/Progenitor Cell Property

In this study, we applied MSC isolation procedures to collect adherent cells rather than free-floating aggregated neurospheres isolated by the standard neuron stem cell method [19]. hSCPs exhibited MSC capacity when the neurogenic cytokines (EGF and FGF-β) were not added to the differentiation media.

The surface antigens we examined were nearly identical between hSCPs and BM-MSCs, except for a higher expression of PDGFR and CD49f in hSCPs. PDGFRα has been proposed to modulate the differentiation of embryonic neuronal stem cells to astrocytes, oligodendrocytes, and glial cells in vitro [19]. Expression of PDGFRα is a marker of oligodendrocyte precursors in rat spinal cord and neuron precursors in cortex [20, 21]. CD49f, also known as alpha6 integrin, is a member of the laminin family. It is expressed on hematopoietic cells and is associated with beta1 integrin in playing an important role in neuronal survival, especially for oligodendrocytes [22, 23]. c-hSCP showed preferential neurogenic differentiation under appropriate conditions. The frequency of c-hSCP–derived neurongenic differentiated cells (80%, Fig. 6C) after induction was significantly higher than the BM-MSCs ( 30%, Fig. 6D), indicating their preservation of high neurogenic differentiation potential. Therefore, the frequencies of PDGFRα and alpha6 integrin expression in both types of MSCs may be related to the cell population with the potential for neurogenesis and oligodendrogenesis under autocrine/paracrine regulation. EGF is essential for the cell growth of neuronal precursors from the embryonic central nerve system and is widely used in routine culture of neuron stem cells and precursors [2426]. However, both hSCPs and BM-MSCs expressed low levels of epithelial growth factor receptor. Because no EGF was present in the culture medium of BM-MSCs, the reason for ligand receptor expression in BM-MSCs might be due to stimulation by endogenous EGF or by a secreted ligand from the original tissue. It may also be considered an indicator of neurogenic capacity. The other cytokine present in hSCP culture, FGF-β, is considered a mitogenic factor produced by stromal cells, endothelial cells, and hematopoietic cells [2729]. It promotes embryonic and hematopoietic stem cell survival and growth under stringent or stress conditions [30]. In addition, FGF-β can induce marrow stroma cells toward neuronal and astrocyte differentiation [31]. Previous studies have not provided direct evidence of FGF-β synergy with EGF to support the MSC feature of hSCPs. The function of combined FGF-β and EGF may be to support hSCP proliferation and maintain their neuronal stem cell capacity.

Furthermore, hSCPs displayed a diverse physiological status depending on the nature of cytokines for survival and growth. This is a typical phenomenon of neuronal stem cells. We observed the expression of mature astrocyte protein gene GFAP of hSCPs in preinduction culture with a transcript level equivalent to cells under another neurogenic stimulation (data not shown). The MAP2 gene was later potently induced during neurogenic differentiation, suggesting that distinct factors are required for the differentiation of different neurons and accessory cells. Expression of the MAP2 transcript was detectable in the hSCPs even in EGF-depleted and FGF-β–depleted conditions, which was not observed in BM-MSCs.

According to these results, we suggest that hSCPs displayed a multilineage differentiating feature of mesenchymal and neuron stem cell property in a cytokine-dependent manner. Our study has provided, for the first time, direct evidence that hSCPs display an MSC phenotype and an ectodermal neurogenic differentiation capacity. These cells were readily obtained and expanded, which may make them of general interest for therapeutic use. In addition, these cells should be useful for future investigations of abnormal skin development or wound repair.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References